A conditional knockout resource for the genome-wide study of mouse gene function

Please download to get full document.

View again

of 8
5 views
PDF
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Document Description
A conditional knockout resource for the genome-wide study of mouse gene function
Document Share
Document Tags
Document Transcript
  ARTICLE  doi:10.1038/nature10163 A conditional knockout resource for thegenome-wide study of mouse genefunction William C. Skarnes 1 , Barry Rosen 1 , Anthony P. West 1 , Manousos Koutsourakis 1 , Wendy Bushell 1 , Vivek Iyer 1 ,Alejandro O. Mujica 1 { , Mark Thomas 1 , Jennifer Harrow 1 , Tony Cox 1 , David Jackson 1 , Jessica Severin 1 { , Patrick Biggs 1 { , Jun Fu 2 ,Michael Nefedov 3 , Pieter J. de Jong 3 , A. Francis Stewart 2 & Allan Bradley 1 Gene targeting in embryonic stem cells has become the principal technology for manipulation of the mouse genome,offeringunrivalledaccuracyinalleledesignandaccesstoconditionalmutagenesis.Tobringtheseadvantagestothewiderresearch community, large-scale mouse knockout programmes are producing a permanent resource of targetedmutations in all protein-coding genes. Here we report the establishment of a high-throughput gene-targeting pipelinefor the generation of reporter-tagged, conditional alleles. Computational allele design, 96-well modular vectorconstruction and high-efficiency gene-targeting strategies have been combined to mutate genes on an unprecedentedscale. So far, more than 12,000 vectors and 9,000 conditional targeted alleles have been produced in highly germline-competent C57BL/6N embryonic stem cells. High-throughput genome engineering highlighted by this study is broadly applicable to rat and human stem cells and provides a foundation for future genome-wide efforts aimed atdeciphering the function of all genes encoded by the mammalian genome. Followingthecompletesequencingofthehumanandmousegenomes,the functional analysis of each of the twenty thousand or so protein-coding genes remains an important goal and a major technical chal-lenge. Several genome-wide mutagenesis strategies have been appliedin the mouse, including ethyl-nitrosourea (ENU) mutagenesis, trans-posonmutagenesis,genetrappingandgenetargeting.Genetrappinginmouseembryonicstem(ES)cells 1,2 hasbeenthemostproductivesofar,providing hundreds of thousands of random insertional mutations inmore than half of the protein-coding genes in the mouse 3–5 . Notably,these ES cell resources can be archived indefinitely and are easily dis-tributed to the scientific community for the purpose of generating knockout mice. However, gene-trap alleles cannot be precisely engi-neered and the strategy favours genes expressed in mouse ES cells.Giventhelimitationsofgenetrapping,itisclearthatthegenerationof a complete set of gene knockouts in the mouse will require theapplication of gene-targeting technology in ES cells 6–8 . Gene targeting can be used to engineer virtually any alteration in the mammaliangenome by homologous recombination in mouse ES cells, from pointmutations to large chromosomal rearrangements 9,10 . Over the past20years, gene targeting has been used to elucidate the function of more than 5,000 mammalian genes. Scaling this technology to theremainder of the genome presents numerous technical challengesand requires the production of targeted ES cells on an unprecedentedscale, beyond the scope of conventional methodologies.ThefirsttargetingpipelineforEScellswasreportedseveralyearsagobefore the completion of the mouse genome sequence (Velocigene) 11 .Bacterial artificial chromosome (BAC)-based targeting vectors wereconstructed to replace the coding sequence of the target gene with a lacZ  reporterandpromoter-drivenselectioncassette.Oligonucleotidesrequired for the construction of targeting vectors by recombineering were based on cDNA sequencessurrounding thetranslation initiationandterminationsignalsofeachtargetgene,thusrequiringnopreviousknowledge of the underlying genomicstructure ofthegene. Ina singlerecombineering step, modified BAC clones were generated with highefficiencyandusedtotargetgenesinEScells.Correctlytargetedevents,which involved the deletion of up to 70-kilobases (kb) of genomicsequence, were identified using a novel high-throughput allele-counting assay. The deletion of large regions of genomic sequence,although effective for eliminating the function of the target gene, canhave unintended consequences on the regulation of adjacent and dis-tant transcriptional units 12,13 .To support and accelerate progress towards the genetic analysis of allmammaliangenes,large-scaleknockoutconsortiawereestablishedin 2006 with the goal of generating a complete resource of reporter-tagged null mutations in C57BL/6mouse ES cells 14 . C57BL/6is one of the best characterized inbred strains, is the reference strain for themouse genome sequence and breed well in the laboratory. Thus, thestudy of mutant alleles in a pure C57BL/6 genetic background isconsidered to be ideal for large-scale phenotyping efforts that willfollow. Highly germline-competent ES cell lines from the C57BL/6N substrain of mice have been established for this project 15–17 . Acommonwebportalprovidinginformationandaccesstotheresourcehas been established 18 , with links to designated repositories for order-ing vectors, ES cell clones and mice.Here we describe a pipeline for the design and mass parallel con-struction of conditional targeting vectors by serial 96-well BACrecombineering and high-throughput gene targeting in C57BL/6 EScells.Ourpipelineisconfiguredtocreateanumberofusefulresourcesen route to the generation of targeted ES cells (Supplementary Fig. 1).Ongoing large-scale production of targeted ES cell lines demonstratesrates of homologous recombination in C57BL/6 ES cells well abovethe historical average. Our pipeline forms the basis for the generation 1 Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1SA, UK.  2 Biotechnologisches Zentrum, TU Dresden, 01062 Dresden, Germany.  3 Children’s HospitalOaklandResearchInstitute,Oakland,California94609,USA. { Presentaddresses:RegeneronPharmaceuticals,Inc.,Tarrytown,NewYork,USA(A.O.M.);RIKENOmicsScienceCenter,YokohamaCity,Japan(J.S.); Hopkirk Institute, Massey University, New Zealand (P.B.). 1 6 J U N E 2 0 1 1 | V O L 4 7 4 | N AT U R E | 3 3 7 Macmillan Publishers Limited. All rights reserved  ©2011  of thousands of   lacZ  -tagged conditional alleles for the EuropeanConditional Mouse Mutatgenesis (EUCOMM) and the NationalInstitutes of Health Knockout Mouse (KOMP) programs as part of the international knockout effort 14 . Computer-assisted design of alleles Conditional alleles permit the analysis of gene function in a tissue-specific or temporal manner during embryonic and postnatal develop-ment 10,19 .Ourconditionalalleleisbasedonthe‘knockout-first’design 20 ,astrategythatcombinestheadvantagesofbothareporter-taggedandaconditional mutation (Fig. 1 and Supplementary Fig. 2). In contrast tostandard conditional designs, the initial unmodified allele is predictedto generate a null allele through splicing to a  lacZ   trapping elementcontained in the targeting cassette. Our trapping cassettes include themouse  En2  splice acceptor and the SV40 polyadenylation sequences,signals that have proven to be highly effective in creating null alleles inmice 2,21 .The knockout-first allele can be easily modified in ES cells or incrossestotransgenic FLP  and cre mice.Conditionalallelesaregeneratedby removal of the gene-trap cassette by Flp recombinase, which revertsthe mutation to wild type, leaving   loxP   sites on either side of a criticalexon. Subsequent exposure to Cre deletes the critical exon to induce aframeshift mutation and trigger nonsense-mediated decay of themutant transcript. Many   cre  transgenic strains are available for thestudyofgenefunctioninspecifictissuesanddevelopmentaltimepoints(see http://www.creline.org ).Typically,  loxP   sites are placed in introns of genes to avoid disrupt-ing normal transcription, processing and translation of the targetgene. The  loxP   and  FRT   sites are positioned to minimize possibleinterference with the splice sites of the critical exon. In some cases,the presence of the recombinase sites may perturb normal splicing patterns 22 . Thiscaveatnotwithstanding,knockout firstalleles arevery useful for proving the causality of gene disruptions and observedphenotypes. Reversion of the phenotype with Flp, or conversely,induction of the phenotype with Cre, rule out potential effects of secondary linked mutations that can arise in cultured ES cells 23 .Furthermore, removal of the FRT-flanked stop cassette is particularly useful for further studies of genes that present heterozygous lethalphenotypes.The vector design process ideally begins with high-quality manualannotation of gene structures 24 . Manual annotation identifies andresolves errors in automated gene predictions and captures all knowntranscript variantsfrom availablemessengerRNA evidence. However,manual annotation of genes is a time-consuming process and provedrate-limiting in our high-throughput pipeline. Although the accuracy ofautomatedgenepredictionisimproving,vectordesignsforEnsemblgene structures must be approached with caution.To assist in the design of conditional alleles, we developed a compu-tational tool to identify oligonucleotide sequences (50-mers) suitablefor recombineering. These sequences are used to insert a selectioncassette and  loxP   site aroundthe critical exon and torecover homolog-ous sequence from theBAC requiredfor gene targeting (Fig.2a). Moregenerally, these computational tools can be applied to any other mam-malianornon-mammaliangenomeforwhichtheconstructionoflargenumbers of recombineered DNA constructs is desired. Each design isdisplayed on the genome browser (Fig. 2b) and manually inspected tochoose the optimal design. Valid designs are selected for the 5 9 -mostcritical exon(s) that is common to all known transcript variants anddisrupts at least 50% of the protein-coding sequence. Designs arerejected if the deleted region contains highly conserved intronicsequence as these elements are likely to correspond to regulatory ele-ments and complicate the interpretation of the mutant phenotype inmice 12,13 .Approximately 40% of protein-coding genes do not fit our designcriteria, most commonly, small transcription units composed of oneor two exons. Genes with alternative 5 9  end transcripts are also prob-lematical. In some cases, it is not possible to remove a single exon orcluster of exons that disrupts all isoforms. These genes have been setasideforotherpartnerswithintheinternationalknockoutconsortiumto generate standard  lacZ  -tagged deletion alleles using, for example,Velocigene technology  11 . Construction of modular targeting vectors Forthegenerationofconditionalgene-targetingvectors,wedevelopedastrategy for high-throughput, serial, liquid BAC recombineering in 96-wellformat(Fig.3)similartothatreportedfortransgeneproduction 25,26 .Weadoptedamodularstrategyfortheconstructionoftargetingvectorsusing recombineering to create Gateway-adapted intermediate vectors(Fig. 4a) that are later assembled into the final targeting constructthrough  in vitro  Gateway reactions (Fig. 4b). For targeting in C57BL/6N ES cells 16 , we made use ofindexed C57BL/6J BAC libraries 27 for theconstruction of targeting vectors.The construction of Gateway-adapted intermediate targeting vec-tors from BACs involves three consecutive recombineering steps:insertion of an attR1/attR2  zeo-pheS  Gateway element upstream of the critical exon (Fig. 3b and Supplementary Fig. 3); insertion of afloxed  kanR  cassette downstream of a critical exon (Fig. 3c); and sub-cloning of the modified region of genomic DNA (8–10kb) into aGateway-adapted plasmid backbone by gap repair (Fig. 3d and Sup-plementary Fig. 3). Heterologous attR3/attR4 sites are included toenable switching of the plasmid backbone to introduce a negativeselection cassette for positive–negative targeting in ES cells. Theexquisite efficiency and nucleotide precision of Red operon-inducedrecombination in bacteria permitted the assembly of DNA constructsin96-wellformatthroughthreeroundsofrecombineeringwithan80%overall efficiency (Supplementary Table 1). This efficiency of vectorproductionreadilyaccommodatestheneedsoftheglobalmousegene-targeting projects thataim to knock out thousands of genes per year 14 . Assembly of the final targeting constructs Gateway technology hasbeensuccessfullyusedfortheconstructionof large-scale genomic resources 28,29 . The use of Gateway technology minimizes the potential for deleterious mutations common to poly-merase chain reaction (PCR)-based cloning methods. We developed aseriesofpromoterlessandpromoter-drivenselectioncassettesflankedby attL1/attL2 sites (Supplementary Fig. 4). To use positive–negativeselectionforgenetargeting  30 ,aplasmidbackbonewasconstructedthatcontains attL3/attL4 Gateway elements and a diphtheria-toxin-A-chain 31 (DTA) expression cassette. Final targeting constructs were  lacZ neo 3  lacZ  13 FRT FRT  loxP loxP CreFlpCretm1atm1btm1ctm1d  loxP 1123132 Figure 1  |  Schematic of the ‘knockout-first’ conditional allele.  The‘knockout-first’ allele (tm1a) contains an IRES: lacZ   trapping cassette and afloxed promoter-driven  neo  cassette inserted into the intron of a gene,disrupting gene function. Flp converts the ‘knockout-first’ allele to aconditional allele (tm1c), restoring gene activity. Cre deletes the promoter-driven selection cassette and floxed exon of the tm1a allele to generate a  lacZ  -tagged allele (tm1b) or deletes the floxed exon of the tm1c allele to generate aframeshiftmutation(tm1d),triggeringnonsensemediateddecayofthe deletedtranscript. RESEARCH ARTICLE 3 3 8 | N AT U R E | V O L 4 7 4 | 1 6 J U N E 2 0 1 1 Macmillan Publishers Limited. All rights reserved  ©2011  assembled invitro inathree-partGatewayreaction(Fig.4b)in96-wellformat and sequence-confirmed across all recombineered junctions.Final targeting vectors were recovered from 95% of the intermediateplasmids (Supplementary Table 1). Thus, the overall efficiency of  vector constructionis 75%and, so far, we have constructed more than12,000 final targeting vectors.Theintermediatevectorsthemselves(Fig.4a)representanimportantmodular resource that can be re-used to generate alternative vectordesignsoradditionalmutantallelesinthefuture.Forexample,targeting cassettes containing specialized reporters,suchasalkaline phosphataseor green fluorescent protein, can be rapidly assembled to providealternative visualization of gene expression. Furthermore, targeting  ~1 kb ~5 kb ~5 kb 50-mers PhaseTargetingcassette3 ′    loxP –1 1 1 0 0 0 0 2 2 –1 LR-PCRoligonucleotides a Blocks G5U D G3GFGR b Ensembl/HavanageneKO alleles3 ′    loxP TargetingcassetteG5 G3 Figure 2  |  Computational design of oligonucleotides for recombineering and LR-PCR genotyping. a  , A critical exon(s) common to all transcript variants(redbox)isidentified.Recombineeringoligonucleotides(50-mers)areidentified by ArrayOligoSelector 46 within pre-defined blocks (G5, U, D, G3) of genomicsequenceforinsertionsofthetargetingcassetteand3 9 loxP  siteandforplasmid rescue of the 5 9  and 3 9  homology arms by gap repair. For LR-PCR genotyping, multiple primers (25 to 30-mers) are then selected from 1-kbblocksofgenomicsequence(GF,GR)outsidethehomologyarms. b ,Displayof conditional alleles on the Ensembl genome browser (Distributed AnnotationSystem (DAS) source 5 KO alleles). A conditional design for the mergedEnsembl/Havana  Rbmx   gene on the reverse strand is shown. a + Ara+ Ara+ Ara 42 °CDay 0Day 2Day 4Day 6Day 7Day 8Transformation ofRed plasmid intoBAC hostInsertion offloxed KancassetteGap repair intoR3/R4 plasmidTransformationinto Cre +   E. coli  Insertion ofR1/R2 Gatewaycassette bcde cat  ForiRed tet  oritsCCTCTOnOffCTZCTZOnOffR1-  pheS/zeo -R2CTZKZKA OnR3- ori/ampR -R4ZA Cre cat  ori ts pBADgbaA 37 °C37 °C30 °C30 °C30 °CBACpSC101CreC  loxP -  kan -  loxP CTZKOff Figure 3  |  Construction of Gateway-adapted intermediate targeting vectorsby 96-well BAC recombineering.  Recombineering steps and elapsed time areshown.  a  , BAC clones, arrayed in 96-well format and electroporated with aplasmid expressing arabinose-inducible Red proteins (pBADgbaA) 47 . b – d , After arabinose induction, cells are electroporated with PCR fragmentscontaining R1-  pheS/zeo -R2 Gateway element ( b ),  loxP-kan-loxP   cassette( c ) and R3-ori/ ampR -R4 subcloning plasmid ( d ).  e , After gap repair, plasmidDNA is prepared and transformed into Cre-expressing bacteria to remove the kanR  cassette, leaving a single  loxP   site downstream of the critical exon.Antibiotics used at each step are: A, ampicillin; C, chloramphenicol; K,kanamycin; T, tetracycline; Z, zeocin. ARTICLE RESEARCH 1 6 J U N E 2 0 1 1 | V O L 4 7 4 | N AT U R E | 3 3 9 Macmillan Publishers Limited. All rights reserved  ©2011   vectors with different selectable markers can be readily constructed toknock out the second allele of genes for functional studies in homo-zygous ES cells. Finally, knock-ins of wild-type and mutant cDNAsprovide an avenue for detailed structure–function studies or to explorehuman variation. Thus, a permanent library of intermediate targeting plasmids willpermitthe further exploitation oftargetingtechnology inthe future. High-throughput ES cell production Toscaletargetingexperimentstohighthroughput,we optimizedelec-troporation conditionsforC57BL/6NES cells 16 in multi-wellcuvettes.HereweaimedtominimizethenumberofcellsandamountofplasmidDNArequiredtoobtainsufficientdrug-resistantcoloniesforscreening (Table 1). After selection, expansion and freezing, most (65%) ES cellclones retained their ability to colonize the germ line of mice 16 .Homologous recombinants generated with targeting vectors areusually identified by Southern blotting. However, this method is notpractical for large-scale screening. Long-range PCR (LR-PCR) is analternative method 32 which is better-suited to high-throughput geno-typing of ES cell clones. We developed a 384-well LR-PCR method toidentify correctly targeted events (Fig. 5). PCR fragments, amplifiedwith gene-specific primers outside the homology arms in combina-tion with primers in the targeting cassette, were sequence-verified. Ingeneral,LR-PCRwasperformedacrossthe3 9 homologyarm.Becausethetargetedclonesaregenotypedatoneend,non-homologouseventswithin the opposite arm will occur in rare cases. Furthermore, mixedclones composed of targeted and non-targeted cells are not detectedby our high-throughput genotyping protocol. For these reasons,further validation of targeted alleles using standard Southern blotassays is highly recommended before use.Owing to frequent crossover events between the selectable markerand3 9 loxP  site,manyofthetargetedEScellcloneslosethe3 9 loxP  siteand cannot be converted to a conditional allele. To distinguishbetween these two alternative products of homologous recombina-tion, LR-PCR products amplified from the 3 9  homology arm weresequenced with a primer at the  loxP   site. Where 3 9  LR-PCR failedtogenerateaproduct,LR-PCRwasperformedacrossthe5 9 homology arm (5 9  LR-PCR). For these cases, the retention of the 3 9  loxP   site wasconfirmed by PCR between the cassette and 3 9  loxP   site. Gene targeting is highly efficient High-throughput gene targeting depends on achieving high targeting efficiencies. For genes expressed in ES cells, a promoterless targeting strategy (referred to as ‘targeted trapping’) 33 has been shown to yieldtargeting efficiencies averaging above 50%. By design, promoterless vectors effectively suppress the recovery of random non-homologousevents in the genome as only insertions in transcribed loci, in thecorrect orientation and reading frame, will confer drug resistance.We electroporated 1,285 different promoterless constructs andobtained targeted clones from nearly half of these constructs withan average targeting efficiency of 50% (Table 1). These data confirmand extend the results of ref. 33, demonstrating that targeted trapping is a highly efficient method for genes expressed in ES cells.Only half of the promoterless targeting vectors were effective inproducing targeted clones. Electroporation of these vectors produced variable numbers of drug-resistant colonies. In general, high colony numberswere predictive ofsuccessful targetingexperiments,whereaslowcolonynumbersusuallyindicatedafailuretotargetthelocus(Sup-plementary Table 2). The success or failure of a construct correlatedwith the number of clones with gene-trap events in the InternationalGeneTrapConsortiumdatabase(SupplementaryTable3).Thus,gene-trapping dataserve asa usefulguide toidentify the subsetof genes thatare amenable to a promoterless targeting strategy  34 . Correlation withclasses of gene was also observed. For instance, targeted trapping waslesseffectivewithsecretedproteinscomparedtonon-secretedproteins,indicating that our cassette designed for trapping secreted proteins(pL1L2_ST, see Supplementary Fig. 4) 35 is not optimal for this classof gene 36 .Given that only half of all genes are expressed at a sufficient level inEScellstosupportatargeted trappingstrategy, we switchedtousing apromoter-driven cassette for positive selection for non-expressedgenescombinedwithnegativeDTAselectiontoselectagainstrandominsertions. We electroporated different positive–negative targeting cassettes and from the analysis of approximately 30 ES cell clonesper unique construct, we recovered targeted events for 80% of geneswithanaveragetargetingefficiencyof35%(Table1;foracompletelistof targeted genes see Supplementary Data). A combination of factorsprobably contribute to our high targeting efficiencies, including theuse of isogenic DNA, relatively long recombineered homology armsand DTA negative selection.Gene targeting is dependent on both the length and the extent of homology between the targeting vector and the target locus 37–39 . Our vectors typically contain 10kb of homology to the endogenous locusand srcinate from a C57BL/6J BAC library. Although the ES cells arederived from the C57BL/6N sub-strain, the Jackson (J) and NIH (N)substrains of C57BL/6 are very closely related 16 , thus our targeting  vectors will have identical sequence with the ES cell genome in thegreat majority of cases. Negative selection was introduced to improvetargeting efficiencies 30,31 . Overall we observed a threefold enrichmentof targeted clones with DTA counter-selection, consistent with pre- vious observations 30,31,40 (Table 1).Inahigh-throughputpipeline,projectsinevitablyfailatoneormoresteps and overall pipeline efficiency depends on effective recovery of  pL3L4_DTA Intermediate vector pL1L2_BactL/RClonaseTargetingcassetteDTA negativeselection cassette5 ′  arm3 ′  arm AsiSI a 5’ homology arm3 ′  homologyarm    zeo :  pheS attR1attR2attR3attR4ori AsiSI  loxP amp 5 ′  homologyarm b   Figure 4  |  Intermediateandfinaltargetingconstructs. a  ,Schematicshowing the structure of the Gateway-adapted intermediate plasmid. A rare AsiSIrestriction site is included in the gap repair plasmid for linearizing the finaltargeting vector before electroporation of ES cells.  b , Assembly of final targeting  vectorsinamulti-Gatewayreaction.SeeSupplementaryFig.4forafulldescriptionof the custom Gateway-adapted plasmids used for vector construction. Table 1  |  Targeting efficiency using promoterless and promoter-driven cassettes Vector type Number ofuniquetargeting vectorsNumber ofsuccessfulelectroporationsNumber ofcolonies * Number ofgenes targetedGenestargeted(%)Targetingefficiency (%)Number ofcoloniesscreened * Number oftargeted clones * Number of targetedclones with3 9  loxP   site * Promoterless 1,285 778 224 621 48 51 24 12 6Promoter 1,811 1,671 348 1,440 80 35 29 10 3.5Promoter ( 2 DTA) 87 87 729 49 56 12 34 4 1 * Average values. RESEARCH ARTICLE 3 4 0 | N AT U R E | V O L 4 7 4 | 1 6 J U N E 2 0 1 1 Macmillan Publishers Limited. All rights reserved  ©2011  these failures. In our experience, most failures are technical in natureand are most efficiently recovered by repeating the procedure.For example, 70% of targeting experiments are rescued after re-electroporation of cells with an alternative preparation of vectorDNA (Supplementary Data). Similarly, re-synthesis of oligonucleo-tides for recombineering or repeating the Gateway reaction recoversamajorityofintermediateandfinaltargetingvectors(datanotshown).Thus, completion of the mutant resource will require iterative roundsof recovery. Whether some genes are refractory to targeting willbecome apparent once all technical issues have been ruled out. Discussion Our targeting pipeline is the major contributor to the internationalmouse knockout programmes that aim to generate  lacZ  -tagged nullmutations ineveryprotein-codinggene in mouse. With the technology describedhere,morethan9,000geneshavebeensuccessfullytargetedinC57BL/6N ES cells to date. The value of our knockout ES cell resourcecritically depends on the germline potential of individual targetedC57BL/6N ES cell clones. In a separate study  16 , hundreds of targetedcelllines generated in our pipeline were assessed forcontribution tothegermlineafter blastocyst injection.At least65% oftargeted clones colo-nized the germ line of chimaeric mice. Thus, our library of mutantC57BL/6N EScells isrobust and willsupport theproduction of mutantmice for future large-scale phenotyping programmes.The scale of mass parallel vector construction and gene targeting described here has implications for functional genomics and proteo-mics in many model systems. New systematic, genome-scale pro-grammes can now be contemplated. Using available BAC or fosmidgenomeresources,thehigh-throughputproductionofcomplextrans-genes and/or targeting constructs will facilitate the generation of sophisticated, physiologically accurate, cell and animal models. Forexample, tagging all proteins in the mouse genome by knock-in target-ing to establish a proteomic mapping programme equivalent to thehighly successful yeast TAP-tagging programmes 41 is now feasible.In the coming years, it is likely that the genome engineering tech-nologiespioneeredinthemousewillbealsoapplicabletoothermodelsystems such as the rat 42,43 and human pluripotent stem cells 44,45 . Thecapacity for fluent gene targeting also permits the systematic genera-tion of doubly targeted ES cell lines for functional studies by con-ditional mutagenesis, which will serve to complement and extendRNA interference studies by providing complete genetic knockouts.Coupled with the power to differentiate ES cells into many cell types,such resources will not only provide means to gaining unique func-tional insights but will also reduce animal experimentation. With pio-neering methodologies, we have overcome the considerable technicalchallenges involved in establishing the most complex and accuratehigh-throughput functional genomics platform yet attempted. Webelievethatourworkraisesthestandardsofachievementandexpecta-tion for future genome-scale programmes. METHODS SUMMARY Geneannotation and vectordesign software. Manualannotationofmousegenestructureswascarriedoutaspreviouslydescribed 24 .Vectordesignsarebasedonthecurrent release of the Ensembl and Vega databases (NCBIM37 assembly). Criticalexon(s) for each target gene are identified computationally (start phase 2 endphase 5 0). Using ArrayOligoSelector 46 , our software returns a set of six 50-meroligonucleotides at defined distances from the critical exon(s) for recombineering. 96-well recombineering and three-way Gateway reactions.  BAC clones areordered from indexed C57BL/6J libraries 27 (RP23/24), arrayed in 96-well platesand transformed with pBADgbaA 47 plasmid encoding lambda Red recombina-tion proteins. Three rounds of recombineering are carried out serially in 96-wellcultures using DNA cassettes amplified by PCR with primers containing 50-nucleotide homology to target sequences 25,26 . After gap repair, plasmid DNA istransformed into Cre-expressing bacteria to reduce the floxed  kanR  cassette to asingle  loxP   site.Three-way Gateway reactions containing intermediate vector, attL1/attL2 tar-geting cassette and attL3/attL4 DTA plasmids are incubated with LR Clonase IIPlus (Invitrogen), transformed into bacteria and selected on agar plates contain-ing appropriate antibiotics and 4-chlorophenylalanine 48 . Final targeting con-structs are sequence-verified across each recombineered junction, linearizedwith AsiSI and visualized on E-Gels (Invitrogen) to verify their size (Supplemen-tary Fig. 5). ElectroporationofEScellsandLR-PCRgenotyping. ElectroporationofC57BL/6N mouse ES cells 16 with linearized plasmid DNA was carried out in 25-wellelectroporationcuvettes(BTXHarvardApparatus).Stablecloneswereselectedinmedium containing Geneticin (Invitrogen). Typically 32 clones are picked,expanded in 96-well plates and archived in 96-well cryovials (Matrix).Long-range PCR reactionsusing SequalPrep (Life Technologies) or LongAMP(NEB) were carried out with genomic DNA from direct lysis of ES cells grown in96-well plates. PCR products were visualized on E-gels (Supplementary Fig. 6)then treated with exonuclease I and phosphatase (NEB) and sequenced. Full Methods  andanyassociated references areavailableintheonlineversionofthe paperatwww.nature.com/nature. Received 3 August 2009; accepted 27 April 2011. 1. Gossler,A.,Joyner,A.L., Rossant,J.&Skarnes,W.C.Mouseembryonicstemcellsandreporterconstructstodetectdevelopmentallyregulatedgenes. Science 244, 463–465 (1989).2. Skarnes,W.C.,Auerbach,B.A.&Joyner,A.L.Agenetrapapproachinmouseembryonicstemcells:the lacZ  reportedisactivatedbysplicing,reflectsendogenousgeneexpression,andismutagenicinmice. Genes Dev. 6, 903–918(1992).3. Zambrowicz B.P. et al.  Wnk1kinasedeficiency lowersblood pressureinmice: agene-trapscreen toidentifypotentialtargetsfor therapeutic intervention.  Proc.Natl Acad. Sci. USA  100, 14109–14114 (2003).4. InternationalGene TrapConsortium. A publicgene trapresource for mousefunctional genomics.  Nature Genet.  36,  543–544 (2004).5. Hansen,G. M. et al.  Large-scale gene trapping inC57BL/6Nmouse embryonicstem cells.  Genome Res. 18,  1670–1679 (2008).6. Bradley,A., Evans, M., Kaufman, M. H.& Robertson, E.Formation ofgerm-linechimaerasfromembryo-derivedteratocarcinomacelllines. Nature 309, 255–256(1984).7. Robertson,E.,Bradley,A.,Kuehn,M.&Evans,M.Germ-linetransmissionofgenesintroducedintoculturedpluripotential cellsbyretroviral vector.  Nature  323, 445–448 (1986).8. Thomas, K.R.& Capecchi, M.R. Site-directedmutagenesis by gene targeting inmouse embryo-derived stem cells.  Cell  51,  503–512 (1987).  lacZ     neo   FRT FRT  loxP loxP GF5 ′ U 3 ′ U GR LR-PCRprimers5 ′  homology arm3 ′  homology armSequencingprimers3 ′ Us GR 5 ′ Us GFLR LX LR-PCRproducts3 ′  arm Cassette 5 ′  arm  Figure 5  |  Genotyping ES clones by LR-PCR sequencing.  Five LR-PCR reactions are carried out: two 5 9  arm (GF/5 9 U), two 3 9  arm (3 9 U/GR) and onecassette (3 9 U/LX). Sequence verification of LR-PCR products is carried outwith gene-specific primers (GF and GR) and with nested primers in thetargetingcassette(5 9 Usand3 9 Us).Toconfirmthepresenceorabsenceofthe3 9 loxP   site,3 9  armLR-PCRproducts aresequenced witha primer adjacentto the loxP   site (LR). In cases where 3 9  armLR-PCR fails to generate a product, the 3 9 loxP   site is confirmed by sequencing the cassette product. ARTICLE RESEARCH 1 6 J U N E 2 0 1 1 | V O L 4 7 4 | N AT U R E | 3 4 1 Macmillan Publishers Limited. All rights reserved  ©2011
Similar documents
View more...
Search Related
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks